Latent Structure Preserving Hashing

Abstract

Aiming at efficient similarity search, hash functions are designed to embed high-dimensional feature descriptors to low-dimensional binary codes such that similar descriptors will lead to binary codes with a short distance in the Hamming space. It is critical to effectively maintain the intrinsic structure and preserve the original information of data in a hashing algorithm. In this paper, we propose a novel hashing algorithm called Latent Structure Preserving Hashing (LSPH), with the target of finding a well-structured low-dimensional data representation from the original highdimensional data through a novel objective function based on Nonnegative Matrix Factorization (NMF) with their corresponding Kullback-Leibler divergence of data distribution as the regularization term. Via exploiting the joint probabilistic distribution of data, LSPH can automatically learn the latent information and successfully preserve the structure of highdimensional data. To further achieve robust performance with complex and nonlinear data, in this paper, we also contribute a more generalized multi-layer LSPH (ML-LSPH) framework, in which hierarchical representations can be effectively learned by a multiplicative up-propagation algorithm. Once obtaining the latent representations, the hash functions can be easily acquired through multi-variable logistic regression. Experimental results on three large-scale retrieval datasets, i.e., SIFT 1M, GIST 1M and 500 K TinyImage, show that ML-LSPH can achieve better performance than the singlelayer LSPH and both of them outperform existing hashing techniques on large-scale data.

Sequential Compact Code Learning for Unsupervised Image Hashing

“…Due to that basic ECE and RBPL-ECE are both inspired by GP, which is always initialized randomly, all the experiments with our methods have been repetitively car-ried out ten times and the final results shown are the averages of the ten runs with a degree of uncertainty. All of the above methods in our experiments are evaluated on six different lengths of codes (16,32,48,64,80, and 96). MAP of all the algorithms on SIFT 1M and GIST 1M data sets.…”

“…self-taught hashing (STH) [15], latent structure preserving hashing [16], spherical hashing (SpherH) [5], iterative quantization (ITQ) [4], compressed hashing (CH) [17], and so on have also been effectively applied for large-scale data retrieval tasks.…”

Describing Unseen Classes by Exemplars: Zero-Shot Learning Using Grouped Simile Ensemble

“…In [2], Akata et al propose an embedding-based framework that regards all of the defined attributes as a whole representation. Many recent approaches adopt such an embedding manner and achieve promising results [13,4,33,15,7,19,39,8,23]. Besides, similarity-based frameworks also adopt the embedding approach [24,40,41,34,8,25].…”

Towards Fine-Grained Open Zero-Shot Learning: Inferring Unseen Visual Features from Attributes

“…[3,17,31] aim to remove the visual-semantic ambiguity through an intermediate embedding space. [35,4] proposes bilinear joint embeddings to mitigate the distribution difference between visual and semantic spaces. In [5], classifiers of unseen classes are directly estimated by aligning the manifolds of seen classes.…”